Docking studies and development of novel 5-heteroarylamino-2,4-diamino-8-chloropyrimido-[4,5-b]quinolines as potential antimalarials

Article abstract of DOI:10.1016/j.bmcl.2006.02.038

MOE-Dock (Docking software) was used to predict the binding modes of 10 novel and potent 5-substituted amino-2,4-diamino-8-chloropyrimido-[4,5-b]quinolines (compounds I-X) as part of our antimalarial drug development programme. This was done by analyzing the interaction of these compounds with the active sites of 11 enzymes present in Plasmodium falciparum and based on this, effective binding was observed to enzyme P. falciparum glutathione reductase (PfGR). The binding scores for compounds I-X with PfGR were also congruent with their antimalarial activity. Three additional analogs were then designed and synthesized based on the above docking study and the pharmacophoric requirements for this class.

Full text of DOI:10.1016/j.bmcl.2006.02.038

Bioorganic & Medicinal Chemistry Letters 16 (2006) 2613–2617
Docking studies and development of novel
5-heteroarylamino-2,4-diamino-8-chloropyrimido-[4,5-b]quinolines
as potential antimalarials
Advait A. Joshi and C. L. Viswanathan*
Department of Pharmaceutical Chemistry, Bombay College of Pharmacy, Kalina, Santacruz (E), Mumbai 400 098, India
Received 22 December 2005; revised 31 January 2006; accepted 15 February 2006
Available online 3 March 2006
Abstract—MOE-Dock (Docking software) was used to predict the binding modes of 10 novel and potent 5-substituted ami-
no-2,4-diamino-8-chloropyrimido-[4,5-b]quinolines (compounds I–X) as part of our antimalarial drug development pro-
gramme. This was done by analyzing the interaction of these compounds with the active sites of 11 enzymes present in
Plasmodium falciparum and based on this, effective binding was observed to enzyme P. falciparum glutathione reductase
(PfGR). The binding scores for compounds I–X with PfGR were also congruent with their antimalarial activity. Three addi-
tional analogs were then designed and synthesized based on the above docking study and the pharmacophoric requirements
for this class.
Ó 2006 Elsevier Ltd. All rights reserved.
Malaria is a serious health problem and according to the
recent report of WHO, there are one million deaths and
500 million new cases due to malaria annually, predom-
inantly attributed to Plasmodium falciparum.1 More-
over, drug resistance is a serious problem in malaria
and it can be attributed to the use of single drug (mono-
therapy) for treatment and to the adaptation of the
malarial parasite by developing alternate pathways for
survival. Hence, the present strategy for new drug devel-
opment is directed towards identifying essential enzyme
systems in the parasite and developing potent molecules
to inhibit them.
present in P. falciparum. Based on the results of
this study we developed additional three novel
compounds.
The crystal structures of 11 enzymes from P. falcipa-
rum were obtained from the Protein Data Bank.3 The
enzymes were Dihydrofolate reductase-thymidylate
synthase (DH-TS) [1J3I], P. falciparum antioxidant
protein (PfAOP) [1XIY], glutathione reductase (PfGR)
[1ONF], lactate dehydrogenase (LDH) [1LDG],
S-adenosyl-L-homocysteine
hydrolase
(PfSAHH)
[1V8B], malarial purine phosphoribosyltransferase
(HGPRT) [1CJB], plasmepsin II (PlasII) [1SME], prop-
lasmepsin II (ProplasII) [1PFZ], P. falciparum protein
kinase 5 (PfPK5) [1OB3], glutamate dehydrogenase
(GDH) [2BMA] and Plasmodium falciparum peptide
deformylase (PfPDF) [1RQC]. The PDB code of each
is mentioned in square brackets. All the enzyme struc-
tures were checked for missing atoms, bonds and con-
tacts. Hydrogens were added to the enzyme structures.
Water molecules and bound ligands were manually
deleted. All the computations were carried out on a
Pentium 1.6 GHz workstation, 512 MB memory with
Windows operating system and Molecular Operating
Environment (MOE 2003.02)4 as the computational
software.
In our previous publication,2 we had presented the
development of a novel series of 5-substituted amino-
2,4-diamino-8-chloropyrimido-[4,5-b]quinolines based
on pharmacophoric studies and six out of ten com-
pounds were found to be active when evaluated in vivo
in mice by Rane’s test (Table 1).
In this report, we present the extension of the work
on the above series of compounds wherein we carried
out docking studies with 11 enzymes reported to be
Keywords: MOE-Dock; Pyrimidoquinolines; P. falciparum glutathione
reductase.
*
Corresponding author. Tel.: +91 22 26670871; fax: +91 22
26670816; e-mail addresses: joshiadvait78@yahoo.com;
chelakara_viswanathan@yahoo.com
The above operation was repeated for each identi-
fied enzyme and then the active site was generated
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.02.038

2614
A. A. Joshi, C. L. Viswanathan / Bioorg. Med. Chem. Lett. 16 (2006) 2613–2617
Table 1. Activity profile of the novel substituted pyrimidoquinolines
R
R
N
N
1 NH2
N
Cl
N
NH2
Compound
R
H
R1
H
Remarks
I
Active (at 160 mg/kg)
II
H
Inactive
Inactive
SO2NH2
SO2NH
III
H
N
SO2NH
N
N
IV
V
H
H
Curative (at 20 mg/kg and above)
H3CO
H3CO
N
Inactive
OH
N
VI
H
H
H
Curative (at 20 mg/kg and above)
Active (at 80 mg/kg and above)
OH
N
N
VII
OH
N
VIII
IX
Curative (at 20 mg/kg and above)
Active (at 20 and 40 mg/kg)
Inactive
OH
N
O
–CH3
O
O
X
–C2H5
O
for each enzyme using MOE-Alpha Site Finder.
MOE-Dock Box was generated around the active
site of each enzyme. The dimensions of the docking
box were manipulated so as to accommodate all the
amino acid residues present in the active site. After
construction of a docking box, the least energy con-
formers of each of compounds I–X were docked in
the active site. The energy scores of enzyme–ligand
complexes (Utotal in kcal/mol) were recorded. Along
with the Utotal scores, two additional parameters
(Predicted pKi and Andrews pKi) were measured.
Predicted pKi measures how efficiently and com-
pletely the ligand utilizes the space provided to it
for binding, while Andrews pKi is a measure of
the total space provided by the protein for ligand
binding.5 Additionally, the ability of the ligand to
bind to the active site at 100 nM concentration
was measured.6
The docking energy scores (Utotal) indicated better inter-
action between compounds I–X and enzymes DH-TS,
PfGR and HGPRT (low values, shown in bold face)
as compared to binding with other enzymes (high
values) (Table 2).
Comparison of predicted pKi values with Andrews pKi
values (Table 3) showed enhanced occupancy of binding
space with PfGR enzyme as compared to DH-TS and
HGPRT. This was also evident from the higher pKi
H-bond values obtained with PfGR enzyme as com-
pared to DH-TS and HGPRT (Table 4). Further, the
more potent compounds IV, VI and VIII returned a

A. A. Joshi, C. L. Viswanathan / Bioorg. Med. Chem. Lett. 16 (2006) 2613–2617
Table 2. Docking energies (Utotal) of various enzyme–ligand complexes
2615
Enzyme
Total MOE-Dock energy (Utotal in kcal/mol) for compounds I–X
IV VI VII VIII
47.1 86.1 79.6
I
II
305.8
III
300.4
V
IX
210.3
X
DH-TS
PfAOP
PfGR
178.5
522.1
162.2
150.6
75.9
248.3
418.6
205.9
512.2
845.2
85.4
1523.1
355.9
1258.3
965.2
98.6
2556.6
38.1
1819.2
270.8
1630.1
1013.6
91.5
56665.2
55.7
2653.8
196.3
1475.3
956.5
100.6
302.5
152.0
354.8
867.8
1416.2
60124.4
43.2
4191.2
310.6
1256.3
1320.1
81.2
4653.2
360.2
1289.9
1428.3
82.1
LDH
1123.2
987.6
89.6
1026.5
860.5
84.3
1010.3
842.3
115.0
491.6
160.8
300.9
902.6
1265.9
1089.3
812.2
80.6
PfSAHH
HGPRT
PlasII
175.6
121.9
291.7
546.3
1123.2
182.1
125.6
260.6
718.6
1020.4
560.5
177.2
325.2
830.2
1167.6
485.2
101.9
320.7
856.4
1424.3
456.8
151.8
347.4
811.8
1323.6
466.9
125.1
295.4
891.4
1532.1
261.5
140.2
292.5
720.5
1019.6
298.9
159.2
308.2
792.6
1086.2
ProplasII
PfPK5
GDH
PfPDF
Table 3. Predicted pKi versus Andrews pKi of compounds I–X
Enzyme
Predicted pKi versus Andrews pKi of test compounds I–X
I
II
III
IV
V
VI
VII
VIII
IX
X
DH-TS
PfGR
HGPRT
4.0/7.6
5.6/8.3
2.4/5.4
7.2/7.6
4.2/8.0
2.5/5.1
6.4/7.3
3.8/6.4
2.3/5.4
5.0/7.6
7.2/8.6
3.3/6.5
6.4/7.6
4.9/8.3
2.9/5.7
5.9/7.0
7.9/8.9
2.5/7.5
7.7/8.3
7.1/8.3
2.3/6.7
6.0/7.7
7.5/8.9
2.2/7.5
3.9/7.9
2.6/8.1
2.5/6.5
3.2/7.9
2.1/8.1
2.1/6.5
Table 4. pKi H-bonds of test compounds I–X and dock_100 nM values
Enzyme
pKi H-bonds of test compounds I–X and dock_100 nM values
I
II
III
IV
V
VI
VII
VIII
IX
X
a
DH-TS
PfGR
HGPRT
1.6 [0]
1.4 [0]
0.1 [0]
1.5 [0]
2.1 [0]
1.2 [0]
1.7 [0]
2.3 [0]
1.4 [0]
2.8 [1]
0.6 [0]
1.5 [0]
2.3 [0]
1.1 [0]
—
2.6 [1]
0.4 [0]
1.5 [0]
3.0 [0]
1.5 [0]
2.7 [1]
0.6 [0]
1.2 [0]
2.1 [0]
0.5 [0]
0.6 [0]
1.5 [0]
0.2 [0]
a
a
—
—
a No interaction through H-bonding. ‘0’ in square brackets indicates binding above 100 nM and ‘1’ indicates binding up to 100 nM.
value of ‘1’ when binding up to 100 nM was considered
as compared to ‘0’ obtained with less active/inactive
compounds (Table 4). A representative binding mode
of compound IV with the active site of PfGR is given
in Figure 1.
compound XI (mp 212 °C). Similarly, reaction of 1
with aqueous sodium nitrite solution in glacial acetic
acid led to the synthesis of benzotriazole 5 (mp
99 °C), which was then refluxed with dichloromethane
in the presence of sodium hydride and anhydrous
DMF to yield N1-chloromethylbenzotriazole 6 (mp
131–132 °C), which was then condensed with 4 in
the presence of anhydrous DMF and potassium car-
bonate to yield compound XII (mp 172 °C). Reaction
of thiourea 7 with malononitrile 8 in the presence of
sodium ethoxide and anhydrous ethanol gave 4,6-di-
Assuming that the novel 5-substituted amino-2,4-diami-
no-8-chloropyrimido-[4,5-b]quinolines act as antimalari-
als by binding with enzyme PfGR, additional three
novel analogues (compounds XI–XIII) with heterocyclic
substituents on the 5-amino position (Table 5) were de-
signed based on docking studies with PfGR and the
pharmacophoric requirements for this class. The data
of docking studies for compounds XI–XIII with PfGR
are reported in Table 6.
amino-2-mercaptopyrimidine
9
(mp 310 °C, dec),
which on selective methylation at 0 °C for 3 h using
methyl iodide in sodium hydroxide solution led to
4,6-diamino-2-methylthiopyrimidine 10 (mp 181 °C).
Compound 10 was then condensed with 4 in the
presence of anhydrous DMF to yield compound XIII
(mp 226 °C).
Synthesis of compounds XI–XIII is outlined in
Scheme 1. Reaction of o-phenylenediamine 1 with
glacial acetic acid in HCl led to the formation of
2-methylbenzimidazole 2 (mp 176 °C), which on bro-
mination with NBS in CCl4 led to the formation of
The progress of all the reactions was monitored by
thin layer chromatography (TLC). All the com-
pounds were purified by silica gel column chromato-
graphy and then recrystallized. The spectral
characteristics of the compounds XI–XIII are men-
tioned in Table 5.
2-bromomethylbenzimidazole
3
(mp 191 °C). This
was then condensed with 2,4,5-triamino-8-chloropy-
rimido-[4,5-b]quinoline 4 in the presence of anhy-
drous DMF and potassium carbonate to yield

2616
A. A. Joshi, C. L. Viswanathan / Bioorg. Med. Chem. Lett. 16 (2006) 2613–2617
Figure 1. Binding mode of compound IV with PfGR.
Table 5. Structures and spectral data of compounds XI–XIII
Compound
R
Yield (%)
FT-IR (cmÀ1) and 1H NMR (d in ppm)
FT-IR (KBr): 3400 (N–H stretch, amine), 3120 (C–H stretch, aromatic), 2920, 2851 (C–H
stretch, alkane), 761 (C–Cl stretch)
H
N
XI
63
1H NMR (DMSO-d6): d 2.5 (s, 2H, –NH), 3.3 (br s, 4H, 2 –NH2), 4.0 (s, 2H, –CH2–), 7.1–7.3
(m, 4H, Ar-H), 7.4–7.6 (m, 3H, Ar-H)
N
FT-IR (KBr): 3321 (N–H stretch, amine), 3003 (C–H stretch, aromatic), 2924 (C–H stretch,
alkane), 761 (C–Cl stretch)
N
N
XII
66
69
N
1H NMR (DMSO-d6): d 2.5 (s, 1H, –NH), 3.2 (br s, 4H, 2 –NH2), 4.8 (s, 2H, –CH2–), 7.3–7.5
(m, 4H, Ar-H), 7.6–7.8 (m, 3H, Ar-H)
NH2
N
FT-IR (KBr): 3320 (N–H stretch, amine), 3096 (C–H stretch, aromatic), 1645, 1591 (N–H
bend, amine), 1261 (C–N vibrations, amine), 763 (C–Cl stretch)
XIII
H2N
N
1H NMR (DMSO-d6): d 2.7 (s, 1H, –NH), 3.4 (br s, 8H, 4 –NH2), 7.1–7.3 (m, 3H, Ar-H), 8.0
(s, 1H, Ar-H)
Table 6. Docking studies of compounds XI–XIII with PfGR
and the purification procedure was simple when NBS
was used. Longer reaction time was required during the
synthesis of 6 when sodium hydroxide or potassium car-
bonate was used as bases instead of sodium hydride.
Scores
XI
XII
XIII
Energy, Utotal (kcal/mol)
Predicted pKi versus
Andrews pKi
39.9
41.9
81
5.9/6.0
6.1/6.4
5.1/6.5
In conclusion, it can be stated that docking studies
helped in identifying the enzyme (PfGR) as the target
for binding for the novel and potent substituted
pyrimidoquinolines. This led to the development of
three additional novel 5-heteroarylamino-2,4-diamino-
8-chloropyrimido-[4,5-b]quinolines.
pKi H-bonds and
dock_100 nM affinity
2.3 [1]
2.6 [1]
3.2 [0]
During the halogenation of 2, NBS gave better yields as
compared to the use of bromine solution or sulfuryl chlo-
ride. Moreover, the workup procedure was less tedious

A. A. Joshi, C. L. Viswanathan / Bioorg. Med. Chem. Lett. 16 (2006) 2613–2617
2617
Scheme 1. Synthesis of compounds XI–XIII.
2. Joshi, A. A.; Narkhede, S. S.; Viswanathan, C. L. Bioorg.
Med. Chem. Lett. 2005, 15, 73.
Acknowledgment
3. Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.;
Bhat, T. N.; Weissig, H.; Shindyalov, I. N.; Bourne, P. E.
Nucleic Acids Res. 2000, 28, 235.
The authors thank University Grants Commission
(UGC), New Delhi, for the financial support.
4. Molecular Operating Environment (MOE) 2003.02, Chem-
ical Computing Group Inc., Quebec, Canada, 2003.
5. Andrews, P. R.; Craik, D. J.; Martin, J. L. J. Med. Chem.
1984, 27, 1648.
6. Maginn S. Abstracts of papers, SMi’s 3rd Annual Confer-
ence, The Hatton, London, February 23–24, 2004; SMi
11.40.
References and notes
1. WHO World Malaria Report, 2005, World Health Orga-
Geneva,
nization,
wmr2005).
Switzerland
(http://rbm.who.int/